Ethernet/Bridge Packet Loss?

Discussion in 'Linux Admins, Storage and Virtualization' started by Eric Faden, Feb 17, 2018.

  1. Eric Faden

    Eric Faden Member

    Joined:
    Dec 5, 2016
    Messages:
    97
    Likes Received:
    6
    So I have two servers on the same network. One of them is reporting a large number of packets dropped. Specifically on the RX side. I have eth0/eth1 in vmbr0. I am seeing the drops on both eth0 directly and vmbr0. I have tried to dig into it and this is what I found, but I'm not sure how to fix it.

    Thoughts?

    IFCONFIG
    eth0: flags=4163<UP,BROADCAST,RUNNING,MULTICAST> mtu 1500
    ether 0c:c4:7a:db:cd:06 txqueuelen 1000 (Ethernet)
    RX packets 3798432517 bytes 4603290901414 (4.1 TiB)
    RX errors 0 dropped 432175 overruns 0 frame 0
    TX packets 2734533804 bytes 1389664789195 (1.2 TiB)
    TX errors 0 dropped 0 overruns 0 carrier 0 collisions 0

    vmbr0: flags=4163<UP,BROADCAST,RUNNING,MULTICAST> mtu 1500
    inet 10.0.0.100 netmask 255.255.255.0 broadcast 10.0.0.255
    ether 0c:c4:7a:db:cd:06 txqueuelen 1000 (Ethernet)
    RX packets 5652740168 bytes 5468894875113 (4.9 TiB)
    RX errors 0 dropped 18290359 overruns 0 frame 0
    TX packets 531852439 bytes 421938934642 (392.9 GiB)
    TX errors 0 dropped 0 overruns 0 carrier 0 collisions 0

    ETHTOOL (-S eth0)
    NIC statistics:
    rx_packets: 3798443493
    tx_packets: 2734548051
    rx_bytes: 4618488580038
    tx_bytes: 1400662523773
    rx_broadcast: 2699042
    tx_broadcast: 2079394
    rx_multicast: 34410121
    tx_multicast: 24331681
    multicast: 34410121
    collisions: 0
    rx_crc_errors: 0
    rx_no_buffer_count: 0
    rx_missed_errors: 0
    tx_aborted_errors: 0
    tx_carrier_errors: 0
    tx_window_errors: 0
    tx_abort_late_coll: 0
    tx_deferred_ok: 0
    tx_single_coll_ok: 0
    tx_multi_coll_ok: 0
    tx_timeout_count: 0
    rx_long_length_errors: 0
    rx_short_length_errors: 0
    rx_align_errors: 0
    tx_tcp_seg_good: 167146392
    tx_tcp_seg_failed: 0
    rx_flow_control_xon: 0
    rx_flow_control_xoff: 0
    tx_flow_control_xon: 0
    tx_flow_control_xoff: 0
    rx_long_byte_count: 4618488580038
    tx_dma_out_of_sync: 0
    lro_aggregated: 0
    lro_flushed: 0
    tx_smbus: 1
    rx_smbus: 0
    dropped_smbus: 0
    os2bmc_rx_by_bmc: 0
    os2bmc_tx_by_bmc: 1
    os2bmc_tx_by_host: 0
    os2bmc_rx_by_host: 1
    tx_hwtstamp_timeouts: 0
    rx_hwtstamp_cleared: 0
    rx_errors: 0
    tx_errors: 0
    tx_dropped: 0
    rx_length_errors: 0
    rx_over_errors: 0
    rx_frame_errors: 0
    rx_fifo_errors: 0
    tx_fifo_errors: 0
    tx_heartbeat_errors: 0
    tx_queue_0_packets: 2734548050
    tx_queue_0_bytes: 1389676597656
    tx_queue_0_restart: 0
    rx_queue_0_packets: 3798443494
    rx_queue_0_bytes: 4603292419856
    rx_queue_0_drops: 0
    rx_queue_0_csum_err: 3
    rx_queue_0_alloc_failed: 0
     
    #1
  2. MBastian

    MBastian Member

    Joined:
    Jul 17, 2016
    Messages:
    49
    Likes Received:
    10
    In case ipv6 is diasbled on that machine these are most likely "dropped" ipv6 broadcast packets. Otherwise check for jumbo frames (rx_long_byte_count).
     
    #2
    Last edited: Feb 18, 2018
    gigatexal likes this.

Share This Page